How much does Web Search cost?

Web Search costs $0.01/search on CallMissed. 1 credit = ₹1 = $0.01 USD.

How do I use Web Search via API?

Send a POST request to POST /v1/search with model "auto" and your API key. CallMissed uses the OpenAI-compatible format — just change the base URL and model field.

What is the context window of Web Search?

Web Search supports a N/A token context window with up to N/A output tokens.

Back to all models

Web Search

by CallMissed · Released 2026

Real-time web search for grounding LLM responses with current information. Defaults to Serper web search, with Exa (neural) and Firecrawl also available. Flat rate per query with no per-result surcharge.

Web Search

Context Window

N/A

Parameters

N/A

Max Output

N/A

Overview

Web Search Auto is CallMissed's multi-provider web search service, designed to ground LLM responses with current, real-time information. By default it uses Serper web search for fast, citation-backed results, and can also route to Exa for neural/AI-native search or Firecrawl for full-page retrieval.

The service supports three modes: auto (platform default provider), shorter (concise results optimized for quick grounding), and detailed (comprehensive results with full context and sources). All modes are priced at a flat ₹1 per search with no per-result surcharge, making costs predictable and easy to budget.

Web Search Auto is essential for RAG (Retrieval-Augmented Generation) pipelines, fact-checking workflows, and any application that needs to combine LLM reasoning with current information. The multi-provider design means a query can fall back across providers for resilience while keeping a single, flat per-search price.

Pricing

Metric	Price
Price /search	₹1.0000

1 credit = ₹1 = $0.01 USD. Prices shown from provider; CallMissed passes through with ~35% markup.

Key Highlights

Serper web search by default (Exa + Firecrawl also available)
Flat ₹1 per search — no per-result surcharge
Cited sources for grounded, verifiable answers
Modes: auto, shorter, detailed

Benchmarks

Benchmark	Score	Notes
Providers	3	Serper (SERP) + Exa (neural) + Firecrawl (full-page)
Modes	3	Auto, shorter, detailed
Pricing	Flat ₹1	Per search, no per-result surcharge
Latency	<500ms	Typical search response time

Technical Details

Multi-provider routing: Serper (web search, default) + Exa (neural) + Firecrawl (full-page)
Cross-provider fallback for resilience — same normalized response shape
Three modes: auto (platform default), shorter (concise), detailed (comprehensive)
Flat ₹1 per search — predictable, no per-result surcharges
Designed for RAG pipelines and LLM grounding workflows
Returns structured results with titles, snippets, URLs, and source citations

Strengths

Serper web search by default — fast, citation-backed results
Flat ₹1 pricing — predictable costs with no per-result surcharges
Three modes (auto/shorter/detailed) for different use cases
Essential for RAG pipelines and keeping LLM responses current

Limitations

Search returns cited sources rather than a raw ranked SERP
Results may vary in quality for non-English queries

Use Cases

RAG pipelinesFact-checkingReal-time information retrievalGrounding LLM responses

API Example

curl https://api.callmissed.com/v1/search \
  -H "Authorization: Bearer cm_YOUR_KEY" \
  -d '{"query": "latest AI model releases 2026", "mode": "auto"}'

Endpoint: POST /v1/search · Model ID: auto

Try Web Search now

Get 1000 free API credits on signup. No credit card required.

Start free Read docs